Frequently Asked Questions

Full file conversions are kept as long as your subscription is active, or until you choose to delete them. Sample files are stored for 3 days. One-time top-up files are stored for 30 days. Download what you need before that period — once expired, files are deleted without recovery.

Files are processed in the same format as they are uploaded so the result is in the highest quality possible. No algorithm is perfect, but we tune ours to keep the voice natural, intelligible and free of robotic artefacts — and we keep improving it.

Each package comes with a limit on how many minutes of audio or video you can process. You can split that across as many files as you want, as long as their total combined length stays within the package. For example, with a 10-minute package you can process 3 files of 3 minutes, or 2 files of 5 minutes. If a file is a bit longer than a whole minute, it's rounded up to the next whole number — a 1-minute-28-second file counts as 2 minutes.

On input: any common audio or video container — MP3, WAV, FLAC, M4A, OGG, AAC, AIFF for audio, and MP4, MOV, WEBM for video. The voice track is extracted from the video automatically. On output: a lossless audio file in the same family as your source whenever possible.

No. KeepVocals does one thing: it removes the music and the background from a recording and gives you back a clean voice file. We don't export drums, bass, instrumentals or a music-only track. If you need a karaoke or a multi-stem split, this isn't the right tool.

No. Whether your file is ten seconds or three hours, the flow is the same. Long files take longer to process, but they're not capped.

Only you. Files are uploaded directly from your browser to our storage and are not shared with third parties. They are deleted according to the retention rules above, or whenever you delete them manually.

Yes. Drop in an MP4, MOV or WEBM and we'll extract the audio, clean the voice, and hand you back the result. The original video is not modified.